// Author: Vaclav Slavik
// RCS-ID: $Id$
// Copyright: (c) 1999 Vaclav Slavik
-// Licence: wxWindows Licence
+// Licence: wxWindows licence
/////////////////////////////////////////////////////////////////////////////
#ifndef _WX_HTMLPARS_H_
#define _WX_HTMLPARS_H_
-#ifdef __GNUG__
-#pragma interface
+#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
+#pragma interface "htmlpars.h"
#endif
#include "wx/defs.h"
#include "wx/html/htmltag.h"
#include "wx/filesys.h"
+#include "wx/hash.h"
+#include "wx/fontenc.h"
-class wxHtmlParser;
-class wxHtmlTagHandler;
+class WXDLLIMPEXP_BASE wxMBConv;
+class WXDLLIMPEXP_HTML wxHtmlParser;
+class WXDLLIMPEXP_HTML wxHtmlTagHandler;
+class WXDLLIMPEXP_HTML wxHtmlEntitiesParser;
-//--------------------------------------------------------------------------------
-// wxHtmlParser
-// This class handles generic parsing of HTML document : it scans
-// the document and divide it into blocks of tags (where one block
-// consists of starting and ending tag and of text between these
-// 2 tags.
-//--------------------------------------------------------------------------------
+class wxHtmlTextPieces;
+class wxHtmlParserState;
-class WXDLLEXPORT wxHtmlParser : public wxObject
+
+enum wxHtmlURLType
+{
+ wxHTML_URL_PAGE,
+ wxHTML_URL_IMAGE,
+ wxHTML_URL_OTHER
+};
+
+// This class handles generic parsing of HTML document : it scans
+// the document and divide it into blocks of tags (where one block
+// consists of starting and ending tag and of text between these
+// 2 tags.
+class WXDLLIMPEXP_HTML wxHtmlParser : public wxObject
{
DECLARE_ABSTRACT_CLASS(wxHtmlParser)
public:
- wxHtmlParser() : wxObject(), m_HandlersHash(wxKEY_STRING)
- { m_FS = NULL; m_Cache = NULL; m_HandlersStack = NULL; }
+ wxHtmlParser();
virtual ~wxHtmlParser();
// Sets the class which will be used for opening files
wxFileSystem* GetFS() const { return m_FS; }
+ // Opens file if the parser is allowed to open given URL (may be forbidden
+ // for security reasons)
+ virtual wxFSFile *OpenURL(wxHtmlURLType type, const wxString& url) const;
+
// You can simply call this method when you need parsed output.
// This method does these things:
// 1. call InitParser(source);
virtual void InitParser(const wxString& source);
// This must be called after Parse().
virtual void DoneParser();
+
+ // May be called during parsing to immediately return from Parse().
+ virtual void StopParsing() { m_stopParsing = TRUE; }
// Parses the m_Source from begin_pos to end_pos-1.
// (in noparams version it parses whole m_Source)
void DoParsing(int begin_pos, int end_pos);
- inline void DoParsing() {DoParsing(0, m_Source.Length());};
+ void DoParsing();
+
+ // Returns pointer to the tag at parser's current position
+ wxHtmlTag *GetCurrentTag() const { return m_CurTag; }
// Returns product of parsing
// Returned value is result of parsing of the part. The type of this result
wxString* GetSource() {return &m_Source;}
void SetSource(const wxString& src);
+
+ // Sets HTML source and remebers current parser's state so that it can
+ // later be restored. This is useful for on-line modifications of
+ // HTML source (for example, <pre> handler replaces spaces with
+ // and newlines with <br>)
+ virtual void SetSourceAndSaveState(const wxString& src);
+ // Restores parser's state from stack or returns FALSE if the stack is
+ // empty
+ virtual bool RestoreState();
+
+ // Parses HTML string 'markup' and extracts charset info from <meta> tag
+ // if present. Returns empty string if the tag is missing.
+ // For wxHTML's internal use.
+ static wxString ExtractCharsetInformation(const wxString& markup);
+
+ // Returns entity parser object, used to substitute HTML &entities;
+ wxHtmlEntitiesParser *GetEntitiesParser() const { return m_entitiesParser; }
protected:
+ // DOM structure
+ void CreateDOMTree();
+ void DestroyDOMTree();
+ void CreateDOMSubTree(wxHtmlTag *cur,
+ int begin_pos, int end_pos,
+ wxHtmlTagsCache *cache);
+
// Adds text to the output.
// This is called from Parse() and must be overriden in derived classes.
// txt is not guaranteed to be only one word. It is largest continuous part of text
// (= not broken by tags)
// NOTE : using char* because of speed improvements
- virtual void AddText(const char* txt) = 0;
+ virtual void AddText(const wxChar* txt) = 0;
// Adds tag and proceeds it. Parse() may (and usually is) called from this method.
// This is called from Parse() and may be overriden.
virtual void AddTag(const wxHtmlTag& tag);
protected:
- // source being parsed
- wxString m_Source;
-
- // tags cache, used during parsing.
- wxHtmlTagsCache *m_Cache;
- wxHashTable m_HandlersHash;
+ // DOM tree:
+ wxHtmlTag *m_CurTag;
+ wxHtmlTag *m_Tags;
+ wxHtmlTextPieces *m_TextPieces;
+ size_t m_CurTextPiece;
+ wxString m_Source;
+
+ wxHtmlParserState *m_SavedStates;
+
// handlers that handle particular tags. The table is accessed by
// key = tag's name.
// This attribute MUST be filled by derived class otherwise it would
// m_HandlersList is list of all handlers and it is guaranteed to contain
// only one reference to each handler instance.
wxList m_HandlersList;
+ wxHashTable m_HandlersHash;
+
+ DECLARE_NO_COPY_CLASS(wxHtmlParser)
// class for opening files (file system)
wxFileSystem *m_FS;
// handlers stack used by PushTagHandler and PopTagHandler
wxList *m_HandlersStack;
+
+ // entity parse
+ wxHtmlEntitiesParser *m_entitiesParser;
+
+ // flag indicating that the parser should stop
+ bool m_stopParsing;
};
-
-
-
-//--------------------------------------------------------------------------------
-// wxHtmlTagHandler
-// This class (and derived classes) cooperates with wxHtmlParser.
-// Each recognized tag is passed to handler which is capable
-// of handling it. Each tag is handled in 3 steps:
-// 1. Handler will modifies state of parser
-// (using it's public methods)
-// 2. Parser parses source between starting and ending tag
-// 3. Handler restores original state of the parser
-//--------------------------------------------------------------------------------
-
-class WXDLLEXPORT wxHtmlTagHandler : public wxObject
+// This class (and derived classes) cooperates with wxHtmlParser.
+// Each recognized tag is passed to handler which is capable
+// of handling it. Each tag is handled in 3 steps:
+// 1. Handler will modifies state of parser
+// (using it's public methods)
+// 2. Parser parses source between starting and ending tag
+// 3. Handler restores original state of the parser
+class WXDLLIMPEXP_HTML wxHtmlTagHandler : public wxObject
{
DECLARE_ABSTRACT_CLASS(wxHtmlTagHandler)
{ m_Parser->DoParsing(tag.GetBeginPos(), tag.GetEndPos1()); }
wxHtmlParser *m_Parser;
+
+ DECLARE_NO_COPY_CLASS(wxHtmlTagHandler)
};
+// This class is used to parse HTML entities in strings. It can handle
+// both named entities and &#xxxx entries where xxxx is Unicode code.
+class WXDLLIMPEXP_HTML wxHtmlEntitiesParser : public wxObject
+{
+ DECLARE_DYNAMIC_CLASS(wxHtmlEntitiesParser)
+
+public:
+ wxHtmlEntitiesParser();
+ virtual ~wxHtmlEntitiesParser();
+
+ // Sets encoding of output string.
+ // Has no effect if wxUSE_WCHAR_T==0 or wxUSE_UNICODE==1
+ void SetEncoding(wxFontEncoding encoding);
+
+ // Parses entities in input and replaces them with respective characters
+ // (with respect to output encoding)
+ wxString Parse(const wxString& input);
+
+ // Returns character for given entity or 0 if the enity is unknown
+ wxChar GetEntityChar(const wxString& entity);
+
+ // Returns character that represents given Unicode code
+#if wxUSE_UNICODE
+ wxChar GetCharForCode(unsigned code) { return (wxChar)code; }
+#else
+ wxChar GetCharForCode(unsigned code);
+#endif
+
+protected:
+#if wxUSE_WCHAR_T && !wxUSE_UNICODE
+ wxMBConv *m_conv;
+ wxFontEncoding m_encoding;
+#endif
+
+ DECLARE_NO_COPY_CLASS(wxHtmlEntitiesParser)
+};
#endif
projects / wxWidgets.git / blobdiff